numerical instability error : cellulose type fuel (MW > 100 g/mol)

102 views
Skip to first unread message

AACosta

unread,
Apr 16, 2020, 1:33:02 PM4/16/20
to FDS and Smokeview Discussions
Hello,

we are trying to simulate a fire in the open air that is affecting a house but we are getting a numerical instability error once the fire (burner type) starts (250 s after simulation begins). 

According to the UG, FDS ends by dumping out one final Plot3D file giving a hint as to where the error is occurring within the computational domain, but we cannot see this file in the folder. Should we set anything in the input file?

Moreover, we think that our problem might be due to any of these two reasons (according to UG):

- a sudden change in the heat release rate (our burner ramps in an interval of just 1s from F=0 to F=1 (100% HRRPUA) -> quite fast)

- a high (>100 g/mol) molecular weight fuel molecule that is not in the FDS database, Table 14.1. has been set (approximately, similar to cellulose: C7.423H10O5N0.004). According to the UG, in such cases, we can reduce the molecular weight maintaining the atom ratios. Should we then try it again with a molecule defined as: C3.712H5O2.5N0.002? I understand that these fuels will have the same heat of combustion if they have the same oxygen consumption per unit mass of fuel. However, will this influence any ouput? Temperatures? 

I see in Table 14.3 that n-decane has a MW > 100, how does this compound behave in FDS, then? Could we get numerical instabilities using cellulose (C6H10O5) too?

Many thanks for your help,

Kevin McGrattan

unread,
Apr 16, 2020, 2:28:24 PM4/16/20
to fds...@googlegroups.com
When you load the case in Smokeview, do you see Plot3D files listed?

As for fuels, just divide the C, H, and O by some number to reduce the overall MW by preserve the atom ratios,

dr_jfloyd

unread,
Apr 16, 2020, 2:35:46 PM4/16/20
to FDS and Smokeview Discussions
None of the species in Table 14.1 use the assumption of gamma=1.4 to compute the specific heat, they all use fits to NASA or JANAF data. This is why the statement about 100 g/mol excludes species listed in Table 14.1.

I would suggest normalizing your fuel to a lower molecular weight. This will have no effect on the EPUMO2. Since specific heat is tied to the molecular weight when using the assumption of gamma=1.4, there will be slight differences in temperature in regions where you have high fuel concentrations. That typically only occurs very close to the burner. 

AACosta

unread,
Apr 16, 2020, 5:26:53 PM4/16/20
to FDS and Smokeview Discussions
Yes, I see the Plot3D option for the time step of the numerical instability! However, it seems like I only get a slice of temperatures at a given y, no velocity or pressure. Is this normal? The last time step giving the error shows:

       Time Step   24948   April 16, 2020  05:22:01
       Step Size:    0.357E-05 s, Total Time:     250.72 s
       Pressure Iterations:     10
       Maximum Velocity Error:  0.13E+05 on Mesh  19 at (   6  23   1)
       Maximum Pressure Error:  0.11E+12 on Mesh   7 at (   1  45   4)
       ---------------------------------------------------------------

Thanks for your help.

Kevin McGrattan

unread,
Apr 16, 2020, 6:04:06 PM4/16/20
to fds...@googlegroups.com
Hit the up/down left/right arrows, plus x, y , or z. v for velocity vectors.

--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/a01bf032-b55c-48ab-aef4-4f9a02dd5dab%40googlegroups.com.

Glenn Forney

unread,
Apr 16, 2020, 6:18:59 PM4/16/20
to fds...@googlegroups.com
Also page up down will move horizontal planes

AACosta

unread,
Apr 16, 2020, 7:21:52 PM4/16/20
to FDS and Smokeview Discussions
Ok, thank you! I got it!

I still have doubts about the reaction/molecule issue. In the UG it is stated that:

(page 154) 
"For high molecular weight species, use of the default gamma will result in very low values of the specific heat which can cause issues with the default extinction model. In this case it is recommended that enthalpy are provided. If the FDS is determining enthalpies using this relation, then it is recommended that you check the CHID.out and verify that reasonable specific heat values have been created."

I was not able to see specific heat values in the CHID.out file that was generated during the stopped simulation. Where should I check these spec. heat values exactly?


(page 173)
Which is the reference (bibliography) for the Heat of Combustion value stated here for wood?
&REAC ID = 'wood', FUEL = 'WOOD', HEAT_OF_COMBUSTION=16400, SPEC_ID_NU = 'WOOD','AIR','PRODUCTS_2' NU=-1,-1.02063,1 /

(page 153)
"The heat of combustion can be determined in a couple of ways. One approach is to take the difference in the heats of formation for the products (assuming complete combustion) and the reactants. This is typically how values are tabulated for pure fuels (e.g., one species) in handbooks. This ideal heat of combustion does not account for the SOOT_YIELD or CO_YIELD that occurs in a real fire. Carbon and hydrogen that go to soot and CO rather than CO2 and H2O result in a lower effective heat of combustion. Setting IDEAL=.TRUE. will reduce the HEAT_OF_COMBUSTION based upon the inputs for SOOT_YIELD and CO_YIELD.If EPUMO2 is specified instead of HEAT_OF_COMBUSTION, then the EPUMO2 will not be changed."

I don't understand when I should set IDEAL = .TRUE. . In the example stated in page 173 for wood, the HoC is not taking into account IDEAL = .TRUE., so the default IDEAL = .FALSE. is considered. Does it mean that soot and CO yields are not taken into consideration and the 16400 kJ/kg value refers to an ideal value? Am I misunderstanding something from the pasted paragraph?

(page 172) 
I have better checked the UG and seen that for wood I could use a molecule like the one stated in SFPE Handbook (e.g. CH1.7O0.74N0.002).

Thank you in advance for your help.

dr_jfloyd

unread,
Apr 17, 2020, 7:48:05 AM4/17/20
to FDS and Smokeview Discussions
There is a section in the .out file that provides a summary of the properties for each tracked species defined in the input file.

Example cases in the User's Guide are intended to be examples and not references for value to use in your own simulation.That being said 16400 is the value for Douglas Fir in the SFPE Handbook Table A.39.

You should set IDEAL=.TRUE. when the heat of combustion you are using represents complete combustion with no soot or CO formation. You will need to make that judgement yourself based on how the reference source describes the process via which it determines the heat of combustion.

When IDEAL=.TRUE. FDS subtracts the heat of formation of the soot and CO from the heat of combustion. When IDEAL=.FALSE. FDS does not do this.
Reply all
Reply to author
Forward
Message has been deleted
0 new messages